DonorsChoose¶
DonorsChoose.org receives hundreds of thousands of project proposals each year for classroom projects in need of funding. Right now, a large number of volunteers is needed to manually screen each submission before it's approved to be posted on the DonorsChoose.org website.
Next year, DonorsChoose.org expects to receive close to 500,000 project proposals. As a result, there are three main problems they need to solve:
- How to scale current manual processes and resources to screen 500,000 projects so that they can be posted as quickly and as efficiently as possible
- How to increase the consistency of project vetting across different volunteers to improve the experience for teachers
- How to focus volunteer time on the applications that need the most assistance
The goal of the competition is to predict whether or not a DonorsChoose.org project proposal submitted by a teacher will be approved, using the text of project descriptions as well as additional metadata about the project, teacher, and school. DonorsChoose.org can then use this information to identify projects most likely to need further review before approval.
About the DonorsChoose Data Set¶
The train.csv data set provided by DonorsChoose contains the following features:
| Feature | Description |
|---|---|
project_id |
A unique identifier for the proposed project. Example: p036502 |
project_title |
Title of the project. Examples:
|
project_grade_category |
Grade level of students for which the project is targeted. One of the following enumerated values:
|
project_subject_categories |
One or more (comma-separated) subject categories for the project from the following enumerated list of values:
Examples:
|
school_state |
State where school is located (Two-letter U.S. postal code). Example: WY |
project_subject_subcategories |
One or more (comma-separated) subject subcategories for the project. Examples:
|
project_resource_summary |
An explanation of the resources needed for the project. Example:
|
project_essay_1 |
First application essay* |
project_essay_2 |
Second application essay* |
project_essay_3 |
Third application essay* |
project_essay_4 |
Fourth application essay* |
project_submitted_datetime |
Datetime when project application was submitted. Example: 2016-04-28 12:43:56.245 |
teacher_id |
A unique identifier for the teacher of the proposed project. Example: bdf8baa8fedef6bfeec7ae4ff1c15c56 |
teacher_prefix |
Teacher's title. One of the following enumerated values:
|
teacher_number_of_previously_posted_projects |
Number of project applications previously submitted by the same teacher. Example: 2 |
* See the section Notes on the Essay Data for more details about these features.
Additionally, the resources.csv data set provides more data about the resources required for each project. Each line in this file represents a resource required by a project:
| Feature | Description |
|---|---|
id |
A project_id value from the train.csv file. Example: p036502 |
description |
Desciption of the resource. Example: Tenor Saxophone Reeds, Box of 25 |
quantity |
Quantity of the resource required. Example: 3 |
price |
Price of the resource required. Example: 9.95 |
Note: Many projects require multiple resources. The id value corresponds to a project_id in train.csv, so you use it as a key to retrieve all resources needed for a project:
The data set contains the following label (the value you will attempt to predict):
| Label | Description |
|---|---|
project_is_approved |
A binary flag indicating whether DonorsChoose approved the project. A value of 0 indicates the project was not approved, and a value of 1 indicates the project was approved. |
Notes on the Essay Data¶
-
Prior to May 17, 2016, the prompts for the essays were as follows:
- __project_essay_1:__ "Introduce us to your classroom"
- __project_essay_2:__ "Tell us more about your students"
- __project_essay_3:__ "Describe how your students will use the materials you're requesting"
- __project_essay_4:__ "Close by sharing why your project will make a difference"
-
Starting on May 17, 2016, the number of essays was reduced from 4 to 2, and the prompts for the first 2 essays were changed to the following:
- __project_essay_1:__ "Describe your students: What makes your students special? Specific details about their background, your neighborhood, and your school are all helpful."
- __project_essay_2:__ "About your project: How will these materials make a difference in your students' learning and improve their school lives?"
For all projects with project_submitted_datetime of 2016-05-17 and later, the values of project_essay_3 and project_essay_4 will be NaN.
In [1]:
%matplotlib inline
import warnings
warnings.filterwarnings("ignore")
import sqlite3
import pandas as pd
import numpy as np
import nltk
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics import confusion_matrix
from sklearn import metrics
from sklearn.metrics import roc_curve, auc
from nltk.stem.porter import PorterStemmer
import re
# Tutorial about Python regular expressions: https://pymotw.com/2/re/
import string
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
from nltk.stem.wordnet import WordNetLemmatizer
from gensim.models import Word2Vec
from gensim.models import KeyedVectors
import pickle
from tqdm import tqdm_notebook
import os
from plotly import plotly
import plotly.offline as offline
import plotly.graph_objs as go
offline.init_notebook_mode()
from collections import Counter
1.1 Reading Data¶
In [2]:
project_data = pd.read_csv('train_data.csv')
resource_data = pd.read_csv('resources.csv')
In [3]:
print("Number of data points in train data", project_data.shape)
print('-'*50)
print("The attributes of data :", project_data.columns.values)
In [4]:
print("Number of data points in train data", resource_data.shape)
print(resource_data.columns.values)
resource_data.head(2)
Out[4]:
1.2 Data Analysis¶
In [5]:
# PROVIDE CITATIONS TO YOUR CODE IF YOU TAKE IT FROM ANOTHER WEBSITE.
# https://matplotlib.org/gallery/pie_and_polar_charts/pie_and_donut_labels.html#sphx-glr-gallery-pie-and-polar-charts-pie-and-donut-labels-py
# Return the number of each classes present in train_data
y_value_counts = project_data['project_is_approved'].value_counts()
# Converting count into percentage with overall train_data
print("Number of projects thar are approved for funding ", y_value_counts[1], ", (", (y_value_counts[1]/(y_value_counts[1]+y_value_counts[0]))*100,"%)")
print("Number of projects thar are not approved for funding ", y_value_counts[0], ", (", (y_value_counts[0]/(y_value_counts[1]+y_value_counts[0]))*100,"%)")
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(aspect="equal"))
recipe = ["Accepted", "Not Accepted"]
data = [y_value_counts[1], y_value_counts[0]]
wedges, texts = ax.pie(data, wedgeprops=dict(width=0.5), startangle=-40)
bbox_props = dict(boxstyle="square,pad=0.3", fc="w", ec="k", lw=0.72)
kw = dict(xycoords='data', textcoords='data', arrowprops=dict(arrowstyle="-"),
bbox=bbox_props, zorder=0, va="center")
for i, p in enumerate(wedges):
ang = (p.theta2 - p.theta1)/2. + p.theta1
y = np.sin(np.deg2rad(ang))
x = np.cos(np.deg2rad(ang))
horizontalalignment = {-1: "right", 1: "left"}[int(np.sign(x))]
connectionstyle = "angle,angleA=0,angleB={}".format(ang)
kw["arrowprops"].update({"connectionstyle": connectionstyle})
ax.annotate(recipe[i], xy=(x, y), xytext=(1.35*np.sign(x), 1.4*y),
horizontalalignment=horizontalalignment, **kw)
ax.set_title("Nmber of projects that are Accepted and not accepted")
plt.show()
1.2.1 Univariate Analysis: School State¶
In [6]:
# Pandas dataframe groupby count, mean: https://stackoverflow.com/a/19385591/4084039
temp = pd.DataFrame(project_data.groupby("school_state")["project_is_approved"].apply(np.mean)).reset_index()
# if you have data which contain only 0 and 1, then the mean = percentage (think about it)
temp.columns = ['state_code', 'num_proposals']
temp.head()
# How to plot US state heatmap: https://datascience.stackexchange.com/a/9620
# scl = [[0.0, 'rgb(242,240,247)'],[0.2, 'rgb(218,218,235)'],[0.4, 'rgb(188,189,220)'],\
# [0.6, 'rgb(158,154,200)'],[0.8, 'rgb(117,107,177)'],[1.0, 'rgb(84,39,143)']]
# data = [ dict(
# type='choropleth',
# colorscale = scl,
# autocolorscale = False,
# locations = temp['state_code'],
# z = temp['num_proposals'].astype(float),
# locationmode = 'USA-states',
# text = temp['state_code'],
# marker = dict(line = dict (color = 'rgb(255,255,255)',width = 2)),
# colorbar = dict(title = "% of pro")
# ) ]
# layout = dict(
# title = 'Project Proposals % of Acceptance Rate by US States',
# geo = dict(
# scope='usa',
# projection=dict( type='albers usa' ),
# showlakes = True,
# lakecolor = 'rgb(255, 255, 255)',
# ),
# )
# fig = go.Figure(data=data, layout=layout)
# offline.iplot(fig, filename='us-map-heat-map')
Out[6]:
In [7]:
# https://www.csi.cuny.edu/sites/default/files/pdf/administration/ops/2letterstabbrev.pdf
temp.sort_values(by=['num_proposals'], inplace=True)
print("States with lowest % approvals")
print(temp.head(5))
print('='*50)
print("States with highest % approvals")
print(temp.tail(5))
In [8]:
#stacked bar plots matplotlib: https://matplotlib.org/gallery/lines_bars_and_markers/bar_stacked.html
def stack_plot(data, xtick, col2='project_is_approved', col3='total'):
ind = np.arange(data.shape[0])
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, data[col3].values)
p2 = plt.bar(ind, data[col2].values)
plt.ylabel('Projects')
plt.title('Number of projects aproved vs rejected')
plt.xticks(ind, list(data[xtick].values))
plt.legend((p1[0], p2[0]), ('total', 'accepted'))
plt.show()
In [9]:
def univariate_barplots(data, col1, col2='project_is_approved', top=False):
# Count number of zeros in dataframe python: https://stackoverflow.com/a/51540521/4084039
temp = pd.DataFrame(project_data.groupby(col1)[col2].agg(lambda x: x.eq(1).sum())).reset_index()
# Pandas dataframe grouby count: https://stackoverflow.com/a/19385591/4084039
temp['total'] = pd.DataFrame(project_data.groupby(col1)[col2].agg({'total':'count'})).reset_index()['total']
temp['Avg'] = pd.DataFrame(project_data.groupby(col1)[col2].agg({'Avg':'mean'})).reset_index()['Avg']
temp.sort_values(by=['total'],inplace=True, ascending=False)
if top:
temp = temp[0:top]
stack_plot(temp, xtick=col1, col2=col2, col3='total')
print(temp.head(5))
print("="*50)
print(temp.tail(5))
In [10]:
univariate_barplots(project_data, 'school_state', 'project_is_approved', False)
Observation¶
- Every state has greater than 80% success rate in approval
- CA have done more proposal projects than the others. Also we observed from that eventhough they had done many projects, 85% of projects as been approved by this state.
- VT state had done only 80 projects (lowest number of projects) and out of these 64 has been appreoved.
1.2.2 Univariate Analysis: teacher_prefix¶
In [11]:
univariate_barplots(project_data, 'teacher_prefix', 'project_is_approved' , top=False)
Observation¶
- People's (Other than teacher and Dr.) project is approved above 84% and also done above 10K projects.
- Teacher with prefix (Mrs and Ms) had done more around around in the 10K number of project than the other with less than 10K projects.
1.2.3 Univariate Analysis: project_grade_category¶
In [12]:
univariate_barplots(project_data, 'project_grade_category', 'project_is_approved', top=False)
Summary : These all project_grade_category had equivalent(more or less) project's approved 83% and also done above 10K projects.
1.2.4 Univariate Analysis: project_subject_categories¶
In [13]:
catogories = list(project_data['project_subject_categories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())
In [14]:
project_data['clean_categories'] = cat_list
project_data.drop(['project_subject_categories'], axis=1, inplace=True)
project_data.head(2)
Out[14]:
In [15]:
univariate_barplots(project_data, 'clean_categories', 'project_is_approved', top=20)
In [16]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['clean_categories'].values:
my_counter.update(word.split())
In [17]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_cat_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_cat_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved category wise')
plt.xticks(ind, list(sorted_cat_dict.keys()))
plt.show()
In [18]:
for i, j in sorted_cat_dict.items():
print("{:20} :{:10}".format(i,j))
Summary : We found that Literacy_language and Math_Science categories are trending fields that people are focus the most projects
1.2.5 Univariate Analysis: project_subject_subcategories¶
In [19]:
sub_catogories = list(project_data['project_subject_subcategories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
sub_cat_list = []
for i in sub_catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp +=j.strip()+" "#" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_')
sub_cat_list.append(temp.strip())
In [20]:
project_data['clean_subcategories'] = sub_cat_list
project_data.drop(['project_subject_subcategories'], axis=1, inplace=True)
project_data.head(2)
Out[20]:
In [21]:
univariate_barplots(project_data, 'clean_subcategories', 'project_is_approved', top=50)
In [22]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['clean_subcategories'].values:
my_counter.update(word.split())
In [23]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
sub_cat_dict = dict(my_counter)
sorted_sub_cat_dict = dict(sorted(sub_cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_sub_cat_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_sub_cat_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved state wise')
plt.xticks(ind, list(sorted_sub_cat_dict.keys()))
plt.show()
In [24]:
for i, j in sorted_sub_cat_dict.items():
print("{:20} :{:10}".format(i,j))
Summary : As we already seen Literacy_language and Math_Science categories have the most trending field where people are most considerate on this field to make project
Literacy is the sub-category of Literacty_language and Mathematics is the sub-category of Math_Science which have most trending as we observed from previous point (which has to be expected)
1.2.6 Univariate Analysis: Text features (Title)¶
In [25]:
#How to calculate number of words in a string in DataFrame: https://stackoverflow.com/a/37483537/4084039
word_count = project_data['project_title'].str.split().apply(len).value_counts()
word_dict = dict(word_count)
word_dict = dict(sorted(word_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(word_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(word_dict.values()))
plt.ylabel('Numeber of projects')
plt.xlabel('Numeber words in project title')
plt.title('Words for each title of the project')
plt.xticks(ind, list(word_dict.keys()))
plt.show()
In [26]:
approved_title_word_count = project_data[project_data['project_is_approved']==1]['project_title'].str.split().apply(len)
approved_title_word_count = approved_title_word_count.values
rejected_title_word_count = project_data[project_data['project_is_approved']==0]['project_title'].str.split().apply(len)
rejected_title_word_count = rejected_title_word_count.values
In [27]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_title_word_count, rejected_title_word_count])
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project title')
plt.grid()
plt.show()
In [28]:
plt.figure(figsize=(10,3))
sns.kdeplot(approved_title_word_count,label="Approved Projects", bw=0.6)
sns.kdeplot(rejected_title_word_count,label="Not Approved Projects", bw=0.6)
plt.legend()
plt.show()
Summary : We observed that for the project title which 4 tends to have more density however, we cannot find the differentation of project's approval (Approved and not approved projects). So, we can't get any information from this.
1.2.7 Univariate Analysis: Text features (Project Essay's)¶
In [29]:
# merge two column text dataframe:
project_data["essay"] = project_data["project_essay_1"].map(str) +\
project_data["project_essay_2"].map(str) + \
project_data["project_essay_3"].map(str) + \
project_data["project_essay_4"].map(str)
In [30]:
approved_word_count = project_data[project_data['project_is_approved']==1]['essay'].str.split().apply(len)
approved_word_count = approved_word_count.values
rejected_word_count = project_data[project_data['project_is_approved']==0]['essay'].str.split().apply(len)
rejected_word_count = rejected_word_count.values
In [31]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_word_count, rejected_word_count])
plt.title('Words for each essay of the project')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project essays')
plt.grid()
plt.show()
In [32]:
plt.figure(figsize=(10,3))
sns.distplot(approved_word_count, hist=False, label="Approved Projects")
sns.distplot(rejected_word_count, hist=False, label="Not Approved Projects")
plt.title('Words for each essay of the project')
plt.xlabel('Number of words in each eassay')
plt.legend()
plt.show()
Summary : We observed that most of the essay which have 200 words have more density than the others for both approved and not approved project and approved project have more at the peak and not approved project have more for 200 words later. We cannot find any information helpful for further processing
1.2.8 Univariate Analysis: Cost per project¶
In [33]:
# we get the cost of the project using resource.csv file
resource_data.head(2)
Out[33]:
In [34]:
# https://stackoverflow.com/questions/22407798/how-to-reset-a-dataframes-indexes-for-all-groups-in-one-step
price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
price_data.head(2)
Out[34]:
In [35]:
# join two dataframes in python:
project_data = pd.merge(project_data, price_data, on='id', how='left')
In [36]:
approved_price = project_data[project_data['project_is_approved']==1]['price'].values
rejected_price = project_data[project_data['project_is_approved']==0]['price'].values
In [37]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_price, rejected_price])
plt.title('Box Plots of Cost per approved and not approved Projects')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Price')
plt.grid()
plt.show()
In [38]:
plt.figure(figsize=(10,3))
sns.distplot(approved_price, hist=False, label="Approved Projects")
sns.distplot(rejected_price, hist=False, label="Not Approved Projects")
plt.title('Cost per approved and not approved Projects')
plt.xlabel('Cost of a project')
plt.legend()
plt.show()
In [39]:
# http://zetcode.com/python/prettytable/
from prettytable import PrettyTable
#If you get a ModuleNotFoundError error , install prettytable using: pip3 install prettytable
x = PrettyTable()
x.field_names = ["Percentile", "Approved Projects", "Not Approved Projects"]
for i in range(0,101,5):
x.add_row([i,np.round(np.percentile(approved_price,i), 3), np.round(np.percentile(rejected_price,i), 3)])
print(x)
Observation¶
- From the plot graph, we can say for sure that most people are spending on projects about less than 1000 cost price.
- From the percentile table, if we observed on 95th percentile, the cost project (for which the project is rejected) has gotten more costlier than the approved project's ones (which is 124% times of approved ones). Not only that 95% percentile, other less than 95% percentile, rejected project's ones had gotten more expense than the approved ones.
1.2.9 Univariate Analysis: teacher_number_of_previously_posted_projects
Please do this on your own based on the data analysis that was done in the above cells
In [40]:
prev_proj_approved = project_data[project_data['project_is_approved']==1]['teacher_number_of_previously_posted_projects'].values
prev_proj_reject = project_data[project_data['project_is_approved']==0]['teacher_number_of_previously_posted_projects'].values
In [41]:
plt.boxplot([prev_proj_approved, prev_proj_reject])
plt.title('Box Plots of Teacher Number of Previous Posted Projects per approved and not approved Projects')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Tracher Number of Previous Posted Projects')
plt.grid()
plt.show()
In [42]:
plt.figure(figsize=(10,3))
sns.distplot(prev_proj_approved, hist=False, label="Approved Projects")
sns.distplot(prev_proj_reject, hist=False, label="Not Approved Projects")
plt.title('Teacher Number Of Previous Posted Projects per approved and not approved Projects')
plt.xlabel('Teacher Number Of Previous Posted Projects')
plt.legend()
plt.show()
In [43]:
x = PrettyTable()
x.field_names = ["Percentile", "Approved Projects", "Not Approved Projects"]
for i in range(0,101,5):
x.add_row([i,np.round(np.percentile(prev_proj_approved,i), 3), np.round(np.percentile(prev_proj_reject,i), 3)])
print(x)
Observation¶
- From the plot graph, we observed that most of the teachers had posted maximum of 10 previously projects which should more dense in that area.
- From the percentile observation, if we observe 90th and 95th percentile, it showing more gaps about 20-30 projects but from that 95th to 100th percentile, there is huge variation gap between them (300-400)
1.2.10 Univariate Analysis: project_resource_summary
Please do this on your own based on the data analysis that was done in the above cells
Check if the presence of the numerical digits in the project_resource_summary effects the acceptance of the project or not. If you observe that presence of the numerical digits is helpful in the classification, please include it for further process or you can ignore it.
In [44]:
summary_text = project_data[project_data['project_is_approved']==1]['project_resource_summary'].values
print('Total number of data row',summary_text.shape[0])
present_numeric = []
for i in summary_text:
txt = i.split()
for j in txt:
if j.isnumeric():
present_numeric.append(i)
break
else:
continue
print('Number of data row that contain numeric value',len(present_numeric))
print('%age of data row that contain numeric value in all project approved {0}%'.format(summary_text.shape[0]/len(present_numeric)))
In [45]:
summary_text = project_data[project_data['project_is_approved']==0]['project_resource_summary'].values
print('Total number of data row',summary_text.shape[0])
present_numeric = []
for i in summary_text:
txt = i.split()
for j in txt:
if j.isnumeric():
present_numeric.append(i)
break
else:
continue
print('Number of data row that contain numeric value',len(present_numeric))
print('%age of data row that contain numeric value in all project not approved {0}%'.format(summary_text.shape[0]/len(present_numeric)))
Observation¶
- A list of text, which is approved project only, has got 9% that contain numeric value.
- A list of text, which is not approved project only, has got 15% that contain numeric value
From the above points, we can say that there is no relation between numeric text to project approval which we can further process
In [46]:
approved_word_count = project_data[project_data['project_is_approved']==1]['project_resource_summary'].str.split().apply(len)
approved_word_count = approved_word_count.values
rejected_word_count = project_data[project_data['project_is_approved']==0]['project_resource_summary'].str.split().apply(len)
rejected_word_count = rejected_word_count.values
In [47]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_word_count, rejected_word_count])
plt.title('Words for each essay of the project')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project essays')
plt.grid()
plt.show()
In [48]:
plt.figure(figsize=(10,3))
sns.distplot(approved_word_count, hist=False, label="Approved Projects")
sns.distplot(rejected_word_count, hist=False, label="Not Approved Projects")
plt.title('Words for each summary text of the project')
plt.xlabel('Number of words in each summary text')
plt.legend()
plt.show()
Summary : We cant find any useful informatio in from this plot
1.3 Text preprocessing¶
1.3.1 Essay Text¶
In [49]:
project_data.head(2)
Out[49]:
In [50]:
# printing some random essays.
print(project_data['essay'].values[0])
print("="*50)
print(project_data['essay'].values[150])
print("="*50)
print(project_data['essay'].values[1000])
print("="*50)
print(project_data['essay'].values[20000])
print("="*50)
print(project_data['essay'].values[99999])
print("="*50)
In [51]:
# https://stackoverflow.com/a/47091490/4084039
import re
def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)
# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [52]:
sent = decontracted(project_data['essay'].values[20000])
print(sent)
print("="*50)
In [53]:
# \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)
In [54]:
#remove spacial character: https://stackoverflow.com/a/5843547/4084039
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
print(sent)
In [55]:
# https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
'won', "won't", 'wouldn', "wouldn't"]
In [56]:
# Combining all the above statemennts
from tqdm import tqdm
preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_essays.append(sent.lower().strip())
In [57]:
# after preprocesing
preprocessed_essays[20000]
Out[57]:
In [58]:
# update dataframe for clean essay and remove old essay
project_data['clean_essay'] = preprocessed_essays
project_data.drop(['essay'], axis=1, inplace=True)
project_data.head(2)
Out[58]:
1.3.2 Project title Text
In [59]:
# similarly you can preprocess the titles also
project_data.head(2)
Out[59]:
In [60]:
#
preprocessed_title = []
for sentance in tqdm(project_data['project_title'].values):
sent = decontracted(sentance)
# Replacing \r, \, \n into space
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
# Removing special characters other than A-Z a-z and 0-9
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_title.append(sent.lower().strip())
In [61]:
# Updating dataframe for clean project title and remove old project title
project_data['clean_project_title'] = preprocessed_title
project_data.drop(['project_title'], axis=1, inplace=True)
project_data.head(2)
Out[61]:
1.3.3 Project Resource Summary¶
In [62]:
#
preprocessed_project_resource_summary = []
for sentance in tqdm(project_data['project_resource_summary'].values):
sent = decontracted(sentance)
# Removing special characters other than A-Z a-z and 0-9
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_project_resource_summary.append(sent.lower().strip())
In [63]:
# Updating dataframe for clean project resource summary and remove old project resource summary
project_data['clean_project_resource_summary'] = preprocessed_project_resource_summary
project_data.drop(['project_resource_summary'], axis=1, inplace=True)
project_data.head(2)
Out[63]:
1. 4 Preparing data for models¶
In [64]:
project_data.columns
Out[64]:
we are going to consider
- school_state : categorical data
- clean_categories : categorical data
- clean_subcategories : categorical data
- project_grade_category : categorical data
- teacher_prefix : categorical data
- project_title : text data (clean)
- text : text data (clean essay)
- project_resource_summary: text data
- quantity : numerical
- teacher_number_of_previously_posted_projects : numerical
- price : numerical1.4.1 Vectorizing Categorical data¶
In [65]:
# we use count vectorizer to convert the values into one hot encoded features
from sklearn.feature_extraction.text import CountVectorizer
vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_categories'].values)
print(vectorizer.get_feature_names())
categories_one_hot = vectorizer.transform(project_data['clean_categories'].values)
print("Shape of matrix after one hot encodig ",categories_one_hot.shape)
In [66]:
# we use count vectorizer to convert the values into one hot encoded features
vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_subcategories'].values)
print(vectorizer.get_feature_names())
sub_categories_one_hot = vectorizer.transform(project_data['clean_subcategories'].values)
print("Shape of matrix after one hot encodig ",sub_categories_one_hot.shape)
In [67]:
# Please do the similar feature encoding with state, teacher_prefix and project_grade_category also
# One hot encoding for school state
# Count Vectorize with vocuabulary contains unique code of school state and we are doing boolen BoW
vectorizer = CountVectorizer(vocabulary=project_data['school_state'].unique(), lowercase=False, binary=True)
vectorizer.fit(project_data['school_state'].values)
print(vectorizer.get_feature_names())
school_state_one_hot = vectorizer.transform(project_data['school_state'].values)
print("Shape of matrix after one hot encodig ",school_state_one_hot.shape)
In [68]:
# One hot encoding for project_grade_category
# Count Vectorize with vocuabulary contains unique code of project_grade_category and we are doing boolen BoW
vectorizer = CountVectorizer(vocabulary=project_data['project_grade_category'].unique(), lowercase=False, binary=True)
vectorizer.fit(project_data['project_grade_category'].values)
print(vectorizer.get_feature_names())
project_grade_category_one_hot = vectorizer.transform(project_data['project_grade_category'].values)
print("Shape of matrix after one hot encodig ",project_grade_category_one_hot.shape)
In [69]:
# One hot encoding for teacher_prefix
# Count Vectorize with vocuabulary contains unique code of teacher_prefix and we are doing boolen BoW
# Since some of the data is filled with nan. So we update the nan to 'None' as a string
project_data['teacher_prefix'] = project_data['teacher_prefix'].fillna('None')
vectorizer = CountVectorizer(vocabulary=project_data['teacher_prefix'].unique(), lowercase=False, binary=True)
vectorizer.fit(project_data['teacher_prefix'].values)
print(vectorizer.get_feature_names())
teacher_prefix_one_hot = vectorizer.transform(project_data['teacher_prefix'].values)
print("Shape of matrix after one hot encodig ",teacher_prefix_one_hot.shape)
1.4.2 Vectorizing Text data¶
1.4.2.1 Bag of words¶
In [70]:
# We are considering only the words which appeared in at least 10 documents(rows or projects).
vectorizer = CountVectorizer(min_df=10)
text_bow = vectorizer.fit_transform(preprocessed_essays)
print("Shape of matrix after one hot encodig ",text_bow.shape)
1.4.2.2 Bag of Words on `project_title`
In [71]:
# you can vectorize the title also
# before you vectorize the title make sure you preprocess it
# Already Preprocessed the project_title in text preprocessing steps
In [72]:
# Similarly you can vectorize for title also
vectorizer = CountVectorizer(min_df=10)
title_bow = vectorizer.fit_transform(preprocessed_title)
print("Shape of matrix after one hot encodig ",title_bow.shape)
1.4.2.3 Bag of Words on `project_resource_summary`
In [73]:
# Similarly you can vectorize for project resource summary also
vectorizer = CountVectorizer(min_df=10)
pr_summary_bow = vectorizer.fit_transform(preprocessed_project_resource_summary)
print("Shape of matrix after one hot encodig ",pr_summary_bow.shape)
1.4.2.3 TFIDF vectorizer¶
In [74]:
from sklearn.feature_extraction.text import TfidfVectorizer
vectorizer = TfidfVectorizer(min_df=10)
text_tfidf = vectorizer.fit_transform(preprocessed_essays)
print("Shape of matrix after one hot encodig ",text_tfidf.shape)
1.4.2.4 TFIDF Vectorizer on `project_title`
In [75]:
# Similarly you can vectorize for title also
vectorizer = TfidfVectorizer(min_df=10)
title_tfidf = vectorizer.fit_transform(preprocessed_title)
print("Shape of matrix after one hot encodig ",title_tfidf.shape)
1.4.2.5 TFIDF Vectorizer on `project_resource_summary`
In [76]:
# Similarly you can vectorize for project resource summary also
vectorizer = TfidfVectorizer(min_df=10)
pr_summary_tfidf = vectorizer.fit_transform(preprocessed_project_resource_summary)
print("Shape of matrix after one hot encodig ",pr_summary_tfidf.shape)
1.4.2.5 Using Pretrained Models: Avg W2V¶
In [77]:
'''
# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039
def loadGloveModel(gloveFile):
print ("Loading Glove Model")
f = open(gloveFile,'r', encoding="utf8")
model = {}
for line in tqdm(f):
splitLine = line.split()
word = splitLine[0]
embedding = np.array([float(val) for val in splitLine[1:]])
model[word] = embedding
print ("Done.",len(model)," words loaded!")
return model
model = loadGloveModel('glove.42B.300d.txt')
# ============================
Output:
Loading Glove Model
1917495it [06:32, 4879.69it/s]
Done. 1917495 words loaded!
# ============================
words = []
for i in preproced_texts:
words.extend(i.split(' '))
for i in preproced_titles:
words.extend(i.split(' '))
print("all the words in the coupus", len(words))
words = set(words)
print("the unique words in the coupus", len(words))
inter_words = set(model.keys()).intersection(words)
print("The number of words that are present in both glove vectors and our coupus", \
len(inter_words),"(",np.round(len(inter_words)/len(words)*100,3),"%)")
words_courpus = {}
words_glove = set(model.keys())
for i in words:
if i in words_glove:
words_courpus[i] = model[i]
print("word 2 vec length", len(words_courpus))
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
import pickle
with open('glove_vectors', 'wb') as f:
pickle.dump(words_courpus, f)
'''
Out[77]:
In [78]:
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
# make sure you have the glove_vectors file
with open('glove_vectors', 'rb') as f:
model = pickle.load(f)
glove_words = set(model.keys())
In [79]:
# average Word2Vec
# compute average word2vec for each review.
avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_vectors.append(vector)
print(len(avg_w2v_vectors))
print(len(avg_w2v_vectors[0]))
1.4.2.6 Using Pretrained Models: AVG W2V on `project_title`
In [80]:
# Similarly you can vectorize for title also
avg_w2v_title = []; # the avg-w2v for each project title is stored in this list
for sentence in tqdm(preprocessed_title): # for each project title
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the project title
for word in sentence.split(): # for each word in a project title
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_title.append(vector)
print(len(avg_w2v_title))
print(len(avg_w2v_title[0]))
1.4.2.7 Using Pretrained Models: AVG W2V on `project_resource_summary`
In [81]:
# Similarly you can vectorize for project resource summary also
avg_w2v_summary = []; # the avg-w2v for each project resource summary is stored in this list
for sentence in tqdm(preprocessed_project_resource_summary): # for each project resource summary
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the project resource summary
for word in sentence.split(): # for each word in a project resource summary
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_summary.append(vector)
print(len(avg_w2v_summary))
print(len(avg_w2v_summary[0]))
1.4.2.7 Using Pretrained Models: TFIDF weighted W2V¶
In [82]:
# S = ["abc def pqr", "def def def abc", "pqr pqr def"]
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_essays)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [83]:
# average Word2Vec
# compute average word2vec for each review.
tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_vectors.append(vector)
print(len(tfidf_w2v_vectors))
print(len(tfidf_w2v_vectors[0]))
1.4.2.9 Using Pretrained Models: TFIDF weighted W2V on `project_title`
In [84]:
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_title)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [85]:
# Similarly you can vectorize for title also
# compute average word2vec for each project title.
tfidf_w2v_title = []; # the avg-w2v for each project title is stored in this list
for sentence in tqdm(preprocessed_title): # for each project title
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the project title
for word in sentence.split(): # for each word in a project title
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_title.append(vector)
print(len(tfidf_w2v_title))
print(len(tfidf_w2v_title[0]))
1.4.2.10 Using Pretrained Models: TFIDF weighted W2V on `project_resource_summary`
In [86]:
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_project_resource_summary)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [87]:
# Similarly you can vectorize for title also
# compute average word2vec for each project title.
tfidf_w2v_summary = []; # the avg-w2v for each project title is stored in this list
for sentence in tqdm(preprocessed_project_resource_summary): # for each project title
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the project title
for word in sentence.split(): # for each word in a project title
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_summary.append(vector)
print(len(tfidf_w2v_summary))
print(len(tfidf_w2v_summary[0]))
1.4.3 Vectorizing Numerical features¶
For price¶
In [88]:
# the cost feature is already in numerical values, we are going to represent the money, as numerical values within the range 0-1
# normalization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
from sklearn.preprocessing import StandardScaler
# price_normalized = standardScalar.fit(project_data['price'].values)
# this will rise the error
# ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ].
# Reshape your data either using array.reshape(-1, 1)
price_scalar = StandardScaler()
price_scalar.fit(project_data['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above mean and variance.
price_normalized = price_scalar.transform(project_data['price'].values.reshape(-1, 1))
In [89]:
price_normalized
Out[89]:
For teacher_number_of_previously_projects¶
In [90]:
# We are going to represent the teacher_number_of_previously_posted_projects, as numerical values within the range 0-1
# normalization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
# teacher_number_of_previously_posted_projects_normalized = standardScalar.fit(project_data['teacher_number_of_previously_posted_projects'].values)
# this will rise the error
# ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ].
# Reshape your data either using array.reshape(-1, 1)
teacher_number_of_previously_posted_projects_scalar = StandardScaler()
teacher_number_of_previously_posted_projects_scalar.fit(project_data['teacher_number_of_previously_posted_projects'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {teacher_number_of_previously_posted_projects_scalar.mean_[0]}, Standard deviation : {np.sqrt(teacher_number_of_previously_posted_projects_scalar.var_[0])}")
# Now standardize the data with above mean and variance.
teacher_number_of_previously_posted_projects_normalized = teacher_number_of_previously_posted_projects_scalar.transform(project_data['teacher_number_of_previously_posted_projects'].values.reshape(-1, 1))
In [91]:
teacher_number_of_previously_posted_projects_normalized
Out[91]:
For quantity¶
In [92]:
# We are going to represent the quantity, as numerical values within the range 0-1
# normalization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
# quantity_normalized = standardScalar.fit(project_data['quantity'].values)
# this will rise the error
# ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ].
# Reshape your data either using array.reshape(-1, 1)
quantity_scalar = StandardScaler()
quantity_scalar.fit(project_data['quantity'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {quantity_scalar.mean_[0]}, Standard deviation : {np.sqrt(quantity_scalar.var_[0])}")
# Now standardize the data with above mean and variance.
quantity_normalized = quantity_scalar.transform(project_data['quantity'].values.reshape(-1, 1))
In [93]:
quantity_normalized
Out[93]:
1.4.4 Merging all the above features¶
we need to merge all the numerical vectors i.e catogorical, text, numerical vectors
- school_state : categorical data - clean_categories : categorical data - clean_subcategories : categorical data - project_grade_category : categorical data - teacher_prefix : categorical data - project_title : text data (clean) - text : text data (clean essay) - project_resource_summary: text data - quantity : numerical - teacher_number_of_previously_posted_projects : numerical - price : numerical
Assignment 2: Apply TSNE
If you are using any code snippet from the internet, you have to provide the reference/citations, as we did in the above cells. Otherwise, it will be treated as plagiarism without citations.
- In the above cells we have plotted and analyzed many features. Please observe the plots and write the observations in markdown cells below every plot.
- EDA: Please complete the analysis of the feature: teacher_number_of_previously_posted_projects
-
- Build the data matrix using these features
- school_state : categorical data (one hot encoding)
- clean_categories : categorical data (one hot encoding)
- clean_subcategories : categorical data (one hot encoding)
- teacher_prefix : categorical data (one hot encoding)
- project_title : text data (BOW, TFIDF, AVG W2V, TFIDF W2V)
- price : numerical
- teacher_number_of_previously_posted_projects : numerical
- Now, plot FOUR t-SNE plots with each of these feature sets.
- categorical, numerical features + project_title(BOW)
- categorical, numerical features + project_title(TFIDF)
- categorical, numerical features + project_title(AVG W2V)
- categorical, numerical features + project_title(TFIDF W2V)
- Concatenate all the features and Apply TNSE on the final data matrix
- Note 1: The TSNE accepts only dense matrices
- Note 2: Consider only 5k to 6k data points to avoid memory issues. If you run into memory error issues, reduce the number of data points but clearly state the number of datat-poins you are using
In [94]:
from sklearn.manifold import TSNE
from scipy.sparse import hstack
2.1 TSNE with `BOW` encoding of `project_title` feature
In [95]:
# please write all of the code with proper documentation and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
print('.....Categorical attributes.....')
print('School State data shape',school_state_one_hot.shape)
print('Categories data shape',categories_one_hot.shape)
print('SubCategory data shape',sub_categories_one_hot.shape)
print('projecy_grade category shape',project_grade_category_one_hot.shape)
print('teacher_prefix data shape',teacher_prefix_one_hot.shape)
print('*'*60)
print('.....Text attributes.....')
# print(text_bow.shape)
print('project title Bow data shape',title_bow.shape)
print('project title TFIDF data shape',title_tfidf.shape)
print('avg weight project title data shape',np.array(avg_w2v_title).shape)
print('tfidf weight project title data shape',np.array(tfidf_w2v_title).shape)
# print(pr_summary_bow.shape)
print('*'*60)
print('.....Numerics attributes.....')
print('price data shape',price_normalized.shape)
print('teacher number of previously project posted',teacher_number_of_previously_posted_projects_normalized.shape)
print('quantity data shape',quantity_normalized.shape)
In [96]:
# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X = hstack((categories_one_hot, sub_categories_one_hot, project_grade_category_one_hot, \
title_bow, \
price_normalized, teacher_number_of_previously_posted_projects_normalized, quantity_normalized))
X.shape
Out[96]:
In [97]:
# To convert sparse to dense array
X = X.toarray()
X.shape
Out[97]:
In [98]:
# Taking 6k data points only
X_s = X[:6000]
X_s.shape
Out[98]:
In [99]:
# Taking class value of 6k data points only
Y = project_data['project_is_approved'].values
Y.shape
Out[99]:
In [100]:
Y_s = Y[:6000]
Y_s.shape
Out[100]:
In [101]:
# Applying TSNE: https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html
X_embedded = TSNE(n_components=2).fit_transform(X_s)
X_embedded.shape
Out[101]:
In [102]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[102]:
In [103]:
df_plot['Label'] = Y_s
df_plot.head()
Out[103]:
In [104]:
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[104]:
In [105]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used BoW) and Perplexity = 30 (Rest Parameter are as default)')
Out[105]:
In [106]:
X_embedded = TSNE(n_components=2, perplexity= 10).fit_transform(X_s)
In [107]:
X_embedded.shape
Out[107]:
In [108]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[108]:
In [109]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[109]:
In [110]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used BoW) and Perplexity = 10 (Rest Parameter are as default)')
Out[110]:
In [111]:
X_embedded = TSNE(n_components=2, perplexity= 10, learning_rate=750).fit_transform(X_s)
X_embedded.shape
Out[111]:
In [112]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[112]:
In [113]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[113]:
In [114]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used BoW) and Perplexity = 10 and Learning Rate = 750 (Rest Parameter are as default)')
Out[114]:
Observation:¶
By using project_title as a BoW, we observed from the graph that there is no separation of cluster between them by changing perplexity and learning rate in above three graphs.
2.2 TSNE with `TFIDF` encoding of `project_title` feature
In [115]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X = hstack((categories_one_hot, sub_categories_one_hot, project_grade_category_one_hot, \
title_tfidf, \
price_normalized, teacher_number_of_previously_posted_projects_normalized, quantity_normalized))
X.shape
Out[115]:
In [116]:
# To convert sparse to dense array
X = X.toarray()
X.shape
Out[116]:
In [117]:
# Taking 6k data points only
X_s = X[:6000]
X_s.shape
Out[117]:
In [118]:
# Applying TSNE: https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html
X_embedded = TSNE(n_components=2).fit_transform(X_s)
X_embedded.shape
Out[118]:
In [119]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[119]:
In [120]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[120]:
In [121]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used TIDF) and Perplexity = 30 (Rest Parameter are as default)')
Out[121]:
In [122]:
X_embedded = TSNE(n_components=2, perplexity= 10).fit_transform(X_s)
X_embedded.shape
Out[122]:
In [123]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[123]:
In [124]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[124]:
In [125]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used TFIDF) and Perplexity = 10 (Rest Parameter are as default)')
Out[125]:
In [126]:
X_embedded = TSNE(n_components=2, perplexity= 10, learning_rate=750).fit_transform(X_s)
X_embedded.shape
Out[126]:
In [127]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[127]:
In [128]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[128]:
In [129]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used TFIDF) and Perplexity = 10 and Learning Rate = 750 (Rest Parameter are as default)')
Out[129]:
Observation¶
From the above graph (while changing hyperparameter like learning rate and perplexity), it is well separated cluster than the BoW features but it still overlapping of two different classes. So we cant find usefulness for further processing
2.3 TSNE with `AVG W2V` encoding of `project_title` feature
In [130]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
X = hstack((categories_one_hot, sub_categories_one_hot, project_grade_category_one_hot, \
avg_w2v_title, \
price_normalized, teacher_number_of_previously_posted_projects_normalized, quantity_normalized))
X.shape
Out[130]:
In [131]:
# To convert sparse to dense array
X = X.toarray()
X.shape
Out[131]:
In [132]:
# Taking 6k data points only
X_s = X[:6000]
X_s.shape
Out[132]:
In [133]:
# Applying TSNE: https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html
X_embedded = TSNE(n_components=2).fit_transform(X_s)
X_embedded.shape
Out[133]:
In [134]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[134]:
In [135]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[135]:
In [136]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used AVG W2V) and Perplexity = 30 (Rest Parameter are as default)')
Out[136]:
In [137]:
X_embedded = TSNE(n_components=2, perplexity= 10).fit_transform(X_s)
X_embedded.shape
Out[137]:
In [138]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[138]:
In [139]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[139]:
In [140]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used AVG W2V) and Perplexity = 10 (Rest Parameter are as default)')
Out[140]:
In [141]:
X_embedded = TSNE(n_components=2, perplexity= 10, learning_rate=750).fit_transform(X_s)
X_embedded.shape
Out[141]:
In [142]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[142]:
In [143]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[143]:
In [144]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used AVG W2V) and Perplexity = 10 and Learning Rate = 750 (Rest Parameter are as default)')
Out[144]:
Observation¶
From the above plot even after changing perplexity and learning rate to get different distinguish plot. However, we observed that it has gotten even worsen than the other two above (BoW and TFIDF features)
2.4 TSNE with `TFIDF Weighted W2V` encoding of `project_title` feature
In [145]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
X = hstack((categories_one_hot, sub_categories_one_hot, project_grade_category_one_hot, \
tfidf_w2v_title, \
price_normalized, teacher_number_of_previously_posted_projects_normalized, quantity_normalized))
X.shape
Out[145]:
In [146]:
# To convert sparse to dense array
X = X.toarray()
X.shape
Out[146]:
In [147]:
# Taking 6k data points only
X_s = X[:6000]
X_s.shape
Out[147]:
In [148]:
# Applying TSNE: https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html
X_embedded = TSNE(n_components=2).fit_transform(X_s)
X_embedded.shape
Out[148]:
In [149]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[149]:
In [150]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[150]:
In [151]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used TFIDF W2V) and Perplexity = 30 (Rest Parameter are as default)')
Out[151]:
In [152]:
X_embedded = TSNE(n_components=2, perplexity= 10).fit_transform(X_s)
X_embedded.shape
Out[152]:
In [153]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[153]:
In [154]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[154]:
In [155]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used TFIDF W2V) and Perplexity = 10 (Rest Parameter are as default)')
Out[155]:
In [156]:
X_embedded = TSNE(n_components=2, perplexity= 10, learning_rate=750).fit_transform(X_s)
X_embedded.shape
Out[156]:
In [157]:
df_plot = pd.DataFrame(data=X_embedded, columns=['Dim_0','Dim_1'])
df_plot.head()
Out[157]:
In [158]:
df_plot['Label'] = Y_s
df_plot['Label'] = df_plot['Label'].replace({0: 'Not Approved', 1: 'Approved'})
df_plot.head()
Out[158]:
In [159]:
# Plot T-SNE: https://towardsdatascience.com/an-introduction-to-t-sne-with-python-example-5a3a293108d1
sns.set_context("notebook", font_scale=1.1)
sns.set_style("ticks")
sns.lmplot(x = 'Dim_0', y = 'Dim_1', data = df_plot, fit_reg=False, legend=True, size=9, hue='Label')
plt.title('T-SNE Plot (When project_title used TFIDF W2V) and Perplexity = 10 and Learning Rate = 750 (Rest Parameter are as default)')
Out[159]:
Observation¶
From the above plot, there still not get any cluster of different class belonging. So this is not going to be helpful information for further process
2.5 Conclusions
Write few sentences about the results that you obtained and the observations you made.¶
- Number of approved project and not approved project of all data points are 85% and 15%
- From the state_code analysis, we observed that CA state_code had done more projects above (15K) and had been approved about 85%. VT state_code had done only 80 projects and out of these, 64 projects had been approved (80%)
- From the teacher_prefix analysis, Teacher with prefix (Mrs and Ms) had done more around around in the 10K number of project than the other with less than 10K projects.
- From the project_grade_category analysis, These all project_grade_category had equivalent(more or less) project's approved 83% (+- 2%)and also done above 10K projects on every grades.
- From the project_subject_categories analysis, We found that Literacy_language and Math_Science categories are trending fields that people are focus the most projects
- From the project_subject_subcategory, As we already seen Literacy_language and Math_Science categories have the most trending field where people are most considerate on this field to make project. Literacy is the sub-category of Literacty_language and Mathematics is the sub-category of Math_Science which have most trending as we observed from previous point (which has to be expected)
- From the project_title analysis, We observed that for the project title which around 4 words tends to have more density however, we cannot find the differentation of project's approval (Approved and not approved projects).
- From the essay, We observed that most of the essay which have 200 words have more density than the others for both approved and not approved project and approved project have more at the peak and not approved project have more for 200 words later.
- From the cost_per_project analysis, we observed from the percentitle table (percentile of 5 interval ) that, rejected projects had spent more expense than the approved one.
- From the teacher_number_of_previously_posted_project analysis, if we observe 90th and 95th percentile, it showing more gaps about 20-30 projects but from that 95th to 100th percentile, there is huge variation gap between them (300-400).
- From the proejct_resource_summary analysis, All list of text, which is approved project only, has got 9% that contain numeric value. and for all list of text, which is not approved project only, has got 15% that contain numeric value. We can say that there is no relation between numeric text to project approval which we can further process
- We performed text preprocessing on essay, project_tile, project_resource_summary like decontract, remove special character, converting to lowercase and removal of stopwords and keeping the clean ones.
- We performed Feature extraction on essay, project_tile, project_resource_summary the text to vector are Bag of Words (BoW), TF-IDF , Average w2v and TF-IDF w2V. And words are taken from pretained from glove vector.
- Converting the numeric features into range value (-1,1) by Standization to neglecence the units.
- Converting the catorgical attibutes into oneHot encode for further processing.
- Applying t-SNE by taking attributes are project_subject_categories, project_subject_subcategory, project_grade_category, project_tile (taking different feature extraction on this attributes only), price, teacher_number_of_previously_posted_project and quantity. We conclude that we didnt get any information even applying this techniques. So, we need to think other powerful technique which can be ddifferentiated from each other classes
In [ ]: